## Saturday, April 28, 2018

### EPA, the nature of regulation, and democracy

My Hoover colleague Richard Epstein posted a revealing essay on the nature of environmental regulation last week, with environmental regulation as a particular example. The contrast with "Environmental Laws Under Siege: Here is why we have them"  in New York Times  and the New Yorker's  Scott Pruitt's Dirty Politics is instructive

Epstein's point is not about the raw amount of or even what's in the regulation, but the procedure by which regulation is imposed:
As drafted, NEPA [National Environmental Policy Act of 1970 ] contains no provision that allows private parties to challenge agency decisions in court. Instead, the NEPA approval process is a matter for internal agency consultation and deliberation that takes into account comments submitted by any interested parties.
One year after its passage, NEPA was turned upside down in a key decision by Judge J. Skelly Wright of the D.C. Circuit Court of Appeals... Wright read the law as giving private parties the right to challenge government actions. Indeed, Wright welcomed such challenges, writing (admiringly) that the change, “promises to become a flood of new litigation—litigation seeking judicial assistance in protecting our natural environment.”
Giving private parties the right to challenge an agency decision grants enormous leverage to the private parties most opposed to letting projects go forward. In the case of nuclear power, delay became the order of the day, as the D.C. Circuit on which Judge Wright sat arrogated to itself the power to find that any EA or EIS was insufficient in some way, so that the entire project was held up until a new and exhaustively updated EIS was prepared—which could then be duly challenged yet again in court.

## Thursday, April 26, 2018

### Conference on the cross section of returns.

The Fama-Miller Center at Chicago Booth jointly with EDHEC and the Review of Financial Studies will host a conference on September 27–28, 2018 in Chicago, on the theme “New Methods for the Cross Section of Returns.” Conference announcement here and call for papers here.

Papers are invited for submission on this broad theme, including:
• Which characteristics provide incremental information for expected returns?
• How can we tame the factor zoo?
• What are the key factors explaining cross-sectional variation in expected returns?
• How many factors do we need to explain the cross section?
• How can we distinguish between competing factor models?
• Do anomaly returns correspond to new factors?
Why a blog post for this among the hundreds of interesting conferences? Naked self-interest. I agreed to give the keynote talk, so the better the conference papers, the more fun I have!

This is, I think, a hot topic, and lots of people are making good progress on it. It's a great time for a conference, and I look forward to catching up and trying to integrate what has been done and were we have to go.

My sense of the topic and the challenge: (Some of this reprises points in "Discount rates," but not all)

Both expected returns and covariances seem to be stable functions of characteristics, like size and book/market ratio. The expected return and covariance of an individual stock seems to vary a lot over time. So we need to build ER(characteristics) and then see if it lines up with covariance(R, factors | characteristics), where factors are also portfolios formed on the basis of characteristics.

## Sunday, April 22, 2018

### Basecoin

Cryptocurrencies like bitcoin have to solve two and a half important problems if they are to become currencies: 1) Unstable values 2) High transactions costs 2.5) Anonymity.

I recently ran across Basis and its Basecoin, an interesting initiative to avoid unstable values. (White paper here.)

Basecoin's idea is to expand and contract the supply so as to maintain a stable value. If the value of the basecoin starts to rise, more will be issued. If it falls, the number will be reduced.

So far so good. But who gets the seignorage when basecoins are increased? And just what do you get for your basecoins if the algorithm is reducing the numbers? From the white paper:
If Basis is trading for more than $1, the blockchain creates and distributes new Basis. These Basis are given by protocol-determined priority to holders of bond tokens and Base Shares, two separate classes of tokens that we’ll detail later. If Basis is trading for less than$1, the blockchain creates and sells bond tokens in an open auction to take coins out of circulation. Bond tokens cost less than 1 Basis, and they have the potential to be redeemed for exactly 1 Basis when Basis is created to expand supply.
Aha, basecoins get traded for ... claims to future basecoins?

You should be able to see instantly how this will unwind. Suppose the algorithm wants to reduce basecoins. It then trades basecoins for "basecoin bonds" which are first-inline promises to receive future basecoin expansions. But those bonds will only have value during temporary drops of demand. If there is a permanent drop in demand, the bonds will never be redeemed and have no value.  They are at best claims to future seignorage. Any peg collapses in a run, and the run threshold is mighty close here.

But it gets worse.

## Thursday, April 19, 2018

### Q is better than you think

This lovely graph comes from "Learning and the Improving Relationship Between Investment and q" by Daniel Andrei, William Mann, and Nathalie Moyen. The careful investment and q measurement make it much better than similar figures I've made for example Figure 4 here. Their paper explores the puzzle, just why did q theory work worse before 1995?

The graph also bears on the "monopoly" debate. Corporations are making huge profits, stocks are high, yet we don't see investment, the story goes -- marginal q must be much less than average q, indicating some sort of fixed factor or rent. Not in the graph.

## Wednesday, April 18, 2018

Two more points occur to me regarding share buybacks. 1)When buybacks increase share prices, and management makes money on that, it's a good thing. The common complaint that buybacks are just a way for managers to enrich themselves is exactly wrong. 2) Maybe it's not so good that banks are buying back shares.  3) The tax bill actually gives incentives against buybacks. What's going on is despite, not because.

Recall the example. A company has $100 in cash, and$100 profitable factory. It has two shares outstanding, each worth $100. The company uses the cash to buy back one share. Now it has one share outstanding, worth$100, and assets of one factory. The shareholders are no wealthier. They used to have $200 in stock. Now they have$100 in stock and $100 in cash. It's a wash. Why do share prices sometimes go up when companies announce buybacks? Well, as before, suppose that management had some zany idea of what to do with the cash that would turn the$100 cash into $80 of value. ("Let's invest in a fleet of corporate Ferraris"). Then the stock would only be worth$180 total, or $90 per share. Buying one share back, even overpaying at$100, raises the other share value from $90 to$100.

That was the big point. Share buybacks are a good way to get money out of firms with no ideas, into firms with good ideas. We want firms to invest, but we don't necessarily want every individual firm to invest. That's the classic fallacy that I think it turning Washington on its head. Best of all we want money going from cash rich old companies to cash starved new companies. Buybacks do that.

1) Management getting rich on buybacks is good.

OK, on to management. Management, buyback critics point out, often has compensation linked to the stock price. They might own stock or own stock options. So when the buyback boosts the stock price, then management gets rich too. Aha! The evil (or so they are portrayed) managers are just doing financial shenanigans to enrich themselves!

The fallacy here, is not stopping to think why the buyback raises the share price in the first place. If it is the main reason given in the finance literature, that this rescues cash that was otherwise going to be mal-invested, then you see the great wisdom of giving management stock options and encouraging them to get rich with buybacks.

## Friday, April 13, 2018

### Fiscal theory of monetary policy

Teaching a PhD class and preparing a few talks led me to a very simple example of an idea, which I'm calling the "fiscal theory of monetary policy." The project is to marry new-Keynesian models, i.e. DSGE models with price stickiness, with the fiscal theory of the price level. The example is simpler than the full analysis with price stickiness in the paper by that title.

It turns out that the FTPL can neatly solve the problems of standard new Keynesian models, and often make very little difference to the actual predictions for time series. This is great news. A new-Keynesian modeler wanting to match some impulse response functions, nervous at the less and less credible underpinnings of new-Keynesian models, can, it appears, just change footnotes about equilibrium selection and get back to work. He or she does not have to throw out a lifetime of work, and start afresh to look at inflation armed with debts and deficits. The interpretation of the model may, however, change a lot.

This is also an extremely conservative (in the non-political sense) approach to curing new-Keynesian model problems. You can keep the entire model, just change some parameter values and solution method, and problems vanish (forward guidance puzzle, frictionless limit puzzle, multiple equilibria at the zero bound, unbelievable off-equilibrium threats etc.) The current NK literature is instead embarked on deep surgery to cure these problems: removing rational expectations, adding constrained or heterogeneous agents, etc. I did not think I would find myself in the strange position trying to save the standard new-Keynesian model, while its developers are eviscerating it! But here we are.

The FTMP model

(From here on in, the post uses Mathjax. It looks great under Chrome, but Safari is iffy. I think I hacked it to work, but if it's mangled, try a different browser. If anyone knows why Safari mangles mathjax and how to fix it let me know.)

Here is the example. The model consists of the usual Fisher equation, $i_{t} = r+E_{t}\pi_{t+1}$ and a Taylor-type interest rate rule $i_{t} = r + \phi \pi_{t}+v_{t}$
$v_{t} =\rho v_{t-1}+\varepsilon_{t}^{i}$
Now we add the government debt valuation equation $\frac{B_{t-1}}{P_{t-1}}\left( E_{t}-E_{t-1}\right) \left( \frac{P_{t-1}% }{P_{t}}\right) =\left( E_{t}-E_{t-1}\right) \sum_{j=0}^{\infty}\frac {1}{R^{j}}s_{t+j}$
Linearizing $$\pi_{t+1}-E_{t}\pi_{t+1}=-\left( E_{t}-E_{t+1}\right) \sum_{j=0}^{\infty }\frac{1}{R^{j}}\frac{s_{t+j}}{b_{t}}=-\varepsilon_{t+1}^{s} \label{unexpi}$$ with $$b=B/P$$. Eliminating the interest rate $$i_{t}$$, the equilibrium of this model is now $$E_{t}\pi_{t+1} =\phi\pi_{t}+v_{t} \label{epi}$$
$\pi_{t+1}-E_{t}\pi_{t+1} =-\varepsilon_{t+1}^{s}$
or, most simply, just $$\pi_{t+1}=\phi\pi_{t}+v_{t}-\varepsilon_{t+1}^{s}. \label{equil_ftmp}$$

Here is a plot of the impulse response function:

## Thursday, April 12, 2018

### Intellectual property

The China trade argument has boiled down to intellectual property and trade. Roughly it has gone like this:
"We need to stop China from selling us all this stuff. Bring the jobs home!"
"Uh, right now the jobs problem is that employers can't find workers. Cheap stuff from China is a boon to American consumers. Tariffs like that on steel cost more steel-using jobs than they save."
"Hm. Ok, but we have to threaten with tariffs to get China to stop requiring our companies to share intellectual property!"
I'm still skeptical about the intellectual property and trade argument. OK, suppose China says that in order for a US company to produce there, it must share intellectual property with a Chinese partner. Just how terrible is this? Just how terrible for the US economy, and society as a whole, justifying a robust policy response -- obviously the company would rather make more profits, but that's not a basis for economic policy.

Intellectual property is different from real property, in that it is nonrival. If you live in my house, I can't live in it. But if you use my equation, my blueprints, my recipe for nanoscale lubricants,  or my designs for specialty oilfield equipment, that does not hamper my use of the same ideas.

Because of this feature, intellectual property is quite different in law, and in economics, than other kinds of property. Ideally, once an idea is produced, it should be distributed freely to everybody. The marginal cost is zero, it is nonrival, so society is best off if everyone gets to use new ideas immediately. Economic growth is the spread of better ideas, and the faster the better. Period.

## Wednesday, April 11, 2018

### Why not taxes?

Reaction to the Washington Post oped (blog post, pdf) on debt  has been sure and swift. We suspected we might get criticized by Republicans for complaining about deficits are a problem.  Instead, the attack came from the left.  Justin Fox hit first, followed by a joint oped by Martin  Baily, Jason Furman, Alan Krueger, Laura Tyson and Janet Yellen. It's almost an official response from the Democratic economic establishment.

Their bottom line, really, is that entitlements and deficits are not a problem. They put the blame pretty much entirely on the recently enacted corporate tax cut.   (I'm simplifying a bit. As did they, a lot.)

By contrast, we focused on entitlement spending -- Social Security, Medicare, Medicaid, VA, pensions, and social programs -- as the central budget problem, and entitlement reform (not "cut") together with a strong focus on economic growth as the best answer. Our warning was that interest costs could rise sharply and unexpectedly and really bring down the party.

Well, deficit equals spending minus tax revenue, so why not just raise taxes to solve the budget problem?

First, let's get a handle on the size and source of the problem.

I. Roughly speaking the long term deficit gap is 5 rising to 10 percentage points of GDP. And the big change is entitlements -- social security, medicare, medicaid, pensions.

For example, even Fox's graph shows social security spending rising from 11% of payroll in 2006 and asymptoting at 18%.

The most recent 2017 CBO long-term budget outlook is quite clear. Long before the tax cut that so upsets our critics was even a glimmer in the President's eye, they were warning of budget problems ahead:
If current laws generally remained unchanged, the Congressional Budget Office projects, ..debt...would reach 150 percent of GDP in 2047. The prospect of such large and growing debt poses substantial risks for the nation....
Why Are Projected Deficits Rising?
In CBO’s projections, deficits rise over the next three decades—from 2.9 percent of GDP in 2017 to 9.8 percent in 2047—because spending growth is projected to outpace growth in revenues (see figure below). In particular, spending as a share of GDP increases for Social Security, the major health care programs (primarily Medicare), and interest on the government’s debt.
The CBO gives us this nice graphs to make the point:

Another CBO's graph follows. Top graph -- where is the spending increase? Social security, health, and interest. Not "other noninterest spending."

(In the bottom graph you see a rosy forecast that individual income taxes will rise a few percent of GDP to help pay for this. Don't be so sure. This comes from inflation pushing us into higher tax brackets and assuming congress won't do anything about it. Notice also how small corporate taxes are in the first place.)

The more recent CBO budget and economic outlook is equally clear: The near term problem is 5 percentage points of GDP:
CBO estimates that the 2018 deficit will total $804 billion....[GDP is$20 Trillion, so that's 4% of GDP]  ... In CBO’s projections, budget deficits continue increasing after 2018, rising from 4.2 percent of GDP this year to 5.1 percent in 2022... Deficits remain at 5.1 percent between 2022 and 2025 ... Over the 2021–2028 period, projected deficits average 4.9 percent of GDP..
Then, things get worse,
In CBO’s projections, outlays for the next three years remain near 21 percent of GDP, which is higher than their average of 20.3 percent over the past 50 years. After that, outlays grow more quickly than the economy does, reaching 23.3 percent of GDP ... by 2028.
That increase reflects significant growth in mandatory spending—mainly because the aging of the population and rising health care costs per beneficiary are projected to increase spending for Social Security and Medicare, among other programs. It also reflects significant growth in interest costs, which are projected to grow more quickly than any other major component of the budget, the result of rising interest rates and mounting debt. ...
And that's only 2028.

You see the problem in our critic's complaint:
"The primary reason the deficit in coming years will now be higher than had been expected is the reduction in tax revenue from last year’s tax cuts, not an increase in spending. This year, revenue is expected to fall below 17 percent of gross domestic product."
Let us take the estimate that the recent tax cut cost $1.5 trillion over 10 years, i.e.$150 billion per year or 0.75% of GDP.  Compared to the $800 billion current deficit it's small potatoes. Compared to the 5 percent to 10 percent of GDP we need to find in the sock drawer, it's peanuts. (Compared to the$10 trillion or more racked up in the last 10 years it's not huge either!)

[Update: Thanks to commenters, I now notice the "had been expected." OK, we expected 4% of GDP deficits, and then they passed a tax cut and now it's 5% of GDP. Sure. On the day that the tax cut was passed, the entire increase in the deficit was due to the tax cut. But our article, and the economy, is about the overall level of the deficit. The problem is what had been expected, not the recent minor change!]

Here is what the CBO has to say about it:
For the next few years, revenues hover near their 2018 level of 16.6 percent of GDP in CBO’s projections. Then they rise steadily, reaching 17.5 percent of GDP by 2025. At the end of that year, many provisions of the 2017 tax act expire, causing receipts to rise sharply—to 18.1 percent of GDP in 2026 and 18.5 percent in 2027 and 2028. They have averaged 17.4 percent of GDP over the past 50 years.
17, maybe 18. We're waddling around in the 1% range, when the problem is in the 10 percent range. The long run budget problem has essentially nothing to do with the Trump tax cut. It has been brewing under Bush, Obama, and Trump. It fundamentally comes from growth in entitlements an order of magnitude larger.

It is simply not true that "The primary reason the deficit in coming years will now be higher than had been expected is the reduction in tax revenue from last year’s tax cuts, not an increase in spending."

To call us "dishonest" -- to call George Shultz "dishonest," in the printed pages of the Washington Post -- for merely repeating what's been in every CBO long term budget forecast for the last two decades really is a new low for economists of this stature. Is Krugmanism infectious?

Put another way, US government debt is about $20 trillion. Various estimates of the entitlement "debt," how much the government has promised more than its revenues, start at$70 trillion and go up in to the hundreds.

To be clear, I agree with the critic's complaint about the tax cut.
"The right way to do reform was to follow the model of the bipartisan tax reform of 1986, when rates were lowered while deductions were eliminated."
Yes! As in many previous blog posts, I am very sad that the chance to do a big 1986 seems to have passed. A large, revenue neutral, distribution neutral, savage cleaning and simplification of the tax code would have been great. There are some elements in the current one -- the lower marginal corporate rate is nice, and there is some capping of deductions, which is why it was a "good first step." But it fell short of my dreams too in many ways.

If only these immensely influential authors had been clamoring for their friends in the Resistance to join forces and pass such a law, rather than (Larry and Jason in particular) spend the whole time arguing that corporate tax cuts just help the rich, perhaps it might have happened. Having to do the whole thing under reconciliation put a lot of limits on what the Republicans could accomplish.

All that aside though, we're still talking about 0.75% of GDP cut compared to a 5%-10% of GDP problem. The long run deficit problem does not come from this tax cut.

II OK, so why not just tax the rich to pay for entitlements?

I hope I have sufficiently dismissed the main line of this particular criticism -- that deficits are all due to the Trump tax cut and all we have to do is put corporate rates back to 35% and all will be well.

On to the larger question, echoed by many commenters on our piece. OK, social security and health are expensive. Let's just tax the rich to pay for it. Like Europe does, so many say.

I do think that roughly speaking we could pay for American social programs with European taxes. That is, 40% payroll taxes rather than our less than 20%; 50% income taxes, starting at very low levels; 20% VAT; various additional taxes like 100% vehicle taxes and gas that costs 3 times ours.

I don't think we can pay for European social programs with European taxes, because Europe can't do it. Their debt/GDP ratios are similar to ours. And their lower growth rates both are the result of this system and compound the problem. Many European countries are responding exactly as we suggest, with deep reforms to their social programs -- less state-paid health insurance, more stringent eligibility requirements and so on.

But that's the option: heavy middle class taxes for middle class benefits, at the cost of substantially lower growth, which itself then drives the needed tax rates up further.

America in fact already has a more progressive tax system than pretty much any other country. Making it more progressive would increase economic distortions dramatically.

A key principle here is that the overall marginal tax rate matters.  There is a tendency, especially on the left, to quote only the top Federal marginal rate of about 40%, and to say therefore that high income Americans pays less taxes than most of Europe. But that argument forgets we also pay state and sometimes local taxes.

The top federal rate is about 40%. In California, we add 13% state income tax, and with no deductibility we're up to 53% right there. But what matters is every wedge between what you produce for your employer and the value of what you get to consume. So we have to add the 7.5% sales tax, so we're up to 60.5% already.

But we're not done.  The Federal corporate tax is now 21%, and California adds 8.84%, so roughly 29% combined. Someone is paying that. If, like sales tax it comes out of higher prices, then add it to the sales tax. Those on the left say no, corporate taxes are all paid by rich people, which is why they were against lowering them. OK, then they contribute fully to the high-income marginal rate.

What about property tax? The main thing people do with a raise in California is to buy a bigger house. Then they pay 1% property tax. As a rough idea, suppose you pay 30% of your income on housing and the price is 20 times the annual cost (typical price/rent ratio). Then you are paying 6% of your income in property taxes. Add 6 percentage points.

I'm not done. All distortions matter. In much of  Europe they charge taxes and then provide people health insurance. We have a cross subsidy scheme, in which you overpay to subsidize others. It's the same as a tax, except much less efficient.  In terms of economic damage, and the overall marginal rate, it should be included. If you live in a condo, whose developer was forced to provide "affordable housing" units, you overpaid just like a tax and a transfer. And so on. I won't try to add these in, but all distortions count.

In sum, we're at a pretty high marginal tax rate already. The notion that we can just blithely raise another 10% of GDP from "the rich" alone without large economic damage does not work.  This isn't a new observation. Just about every study of how to pay for entitlements comes to the same conclusion.

Again, my argument is not about sympathy for the rich. It is a simple cause and effect argument. Marginal tax rates a lot above 70% are going to really damage the economy and not bring in the huge revenue we need.

Bottom line: Paying for the current entitlements entirely by taxes would involve a big tax hike on middle income Americans.

The most important answer is economic growth. 30 years of 3% growth rather than 2% growth gives you 35% more GDP, and thus 35% more tax revenue. If federal revenues are 20% of GDP, that's 7%
of the previous GDP right there.  Deregulation and tax reform -- get on with the lower marginal rates and simplification that we agree on -- are important.

(The CBO also writes,
In CBO’s projections, the effects of the 2017 tax act on incentives to work, save, and invest raise real potential GDP throughout the 2018–2028 period....
The largest effects on GDP over the decade stem from the tax act. In CBO’s projections, it boosts the level of real GDP by an average of 0.7 percent and nonfarm payroll employment by an average of 1.1 million jobs over the 2018–2028 period. During those years, the act also raises the level of real gross national product (GNP) by an annual average of about \$470 per person in 2018 dollars.
This is not a terrible result!)

Our oped was clear to say social program "reform" not just "cut." Little things like changing indexing and retirement ages make a big difference over 30 years. We argue for reducing the growth and expansion of entitlements, not "cut."  Removing some of the very high work disincentives would help people get off some programs. Europe is facing this too, and many countries are a good deal more stringent about qualification than we are.

Our critics say that to point out America cannot pay for the entitlements we have currently promised "dehumanizes the value of these programs to millions of Americans." No. Failing to reform entitlements now and gently will lead to chaotic cuts in the future, on programs that people depend on. If we're going to throw around accusations of heartlessness, denying the problem is the heartless approach.

## Friday, April 6, 2018

### Unraveling

Economists delight in unravelings -- behavioral responses that undo bright ideas. A subsidy for skunks produces cats with white stripes. Two good ones came up this week.

As hare-brained as they are, I have to opine that the actual economic consequences of US steel import tariffs and Chinese soybean tariffs are essentially zero.

(Political comment: tariffs are taxes on imports. It would do fans of the Administration's trade policies good to utter the correct "tax" word to describe tariffs. Or "self-inflicted sanctions." )

Why do I say that? Each country is assessing a tariff on goods produced only by the other country. Well then, why not park the ships overnight in Vancouver, or Tokyo, fill out some paperwork, and say steel is imported first from China to Canada, and then Canada to the U.S., and vice versa?

Trade bureaucrats are smart enough to catch that. But they cannot hope to stop essentially the same thing: China sells steel to Canadian steel users, who currently buy from Canadian firms. Canadian steel producers reorient their production to the US, and sell to US companies who formerly bought from China. The steel is genuinely Canadian.